import numpy as np
import tensorflow.compat.v2 as tf
tf.enable_v2_behavior()
import pandas as pd
from tensorflow import keras
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import RobustScaler
from sklearn.preprocessing import MinMaxScaler
from matplotlib import pyplot
import plotly.graph_objects as go
import math
import seaborn as sns
from sklearn.metrics import mean_squared_error
np.random.seed(1)
tf.random.set_seed(1)
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, LSTM, GRU, Dropout, RepeatVector, TimeDistributed
from keras import backend
MODELFILENAME = 'MODELS/LSTM_3h_TFM_2c'
TIME_STEPS=18 #3h
CMODEL = LSTM
UNITS=45
DROPOUT1=0.118
DROPOUT2=0.243
ACTIVATION='tanh'
OPTIMIZER='adam'
EPOCHS=43
BATCHSIZE=30
VALIDATIONSPLIT=0.2
# Code to read csv file into Colaboratory:
# from google.colab import files
# uploaded = files.upload()
# import io
# df = pd.read_csv(io.BytesIO(uploaded['SentDATA.csv']))
# Dataset is now stored in a Pandas Dataframe
df = pd.read_csv('../../data/dadesTFM.csv')
df.reset_index(inplace=True)
df['Time'] = pd.to_datetime(df['Time'])
df = df.set_index('Time')
columns = ['PM1','PM25','PM10','PM1ATM','PM25ATM','PM10ATM']
df1 = df.copy();
df1 = df1.rename(columns={"PM 1":"PM1","PM 2.5":"PM25","PM 10":"PM10","PM 1 ATM":"PM1ATM","PM 2.5 ATM":"PM25ATM","PM 10 ATM":"PM10ATM"})
df1['PM1'] = df['PM 1'].astype(np.float32)
df1['PM25'] = df['PM 2.5'].astype(np.float32)
df1['PM10'] = df['PM 10'].astype(np.float32)
df1['PM1ATM'] = df['PM 1 ATM'].astype(np.float32)
df1['PM25ATM'] = df['PM 2.5 ATM'].astype(np.float32)
df1['PM10ATM'] = df['PM 10 ATM'].astype(np.float32)
df2 = df1.copy()
train_size = int(len(df2) * 0.8)
test_size = len(df2) - train_size
train, test = df2.iloc[0:train_size], df2.iloc[train_size:len(df2)]
train.shape, test.shape
((3117, 7), (780, 7))
#Standardize the data
for col in columns:
scaler = StandardScaler()
train[col] = scaler.fit_transform(train[[col]])
<ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]]) <ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]]) <ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]]) <ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]]) <ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]]) <ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]])
def create_sequences(X, y, time_steps=TIME_STEPS):
Xs, ys = [], []
for i in range(len(X)-time_steps):
Xs.append(X.iloc[i:(i+time_steps)].values)
ys.append(y.iloc[i+time_steps])
return np.array(Xs), np.array(ys)
X_train, y_train = create_sequences(train[[columns[1]]], train[columns[1]])
#X_test, y_test = create_sequences(test[[columns[1]]], test[columns[1]])
print(f'X_train shape: {X_train.shape}')
print(f'y_train shape: {y_train.shape}')
X_train shape: (3099, 18, 1) y_train shape: (3099,)
#afegir nova mètrica
def rmse(y_true, y_pred):
return backend.sqrt(backend.mean(backend.square(y_pred - y_true), axis=-1))
model = Sequential()
model.add(CMODEL(units = UNITS, return_sequences=True, input_shape=(X_train.shape[1], X_train.shape[2])))
model.add(Dropout(rate=DROPOUT1))
model.add(CMODEL(units = UNITS, return_sequences=True))
model.add(Dropout(rate=DROPOUT2))
model.add(TimeDistributed(Dense(1,kernel_initializer='normal',activation=ACTIVATION)))
model.compile(optimizer=OPTIMIZER, loss='mae',metrics=['mse',rmse])
model.summary()
Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= lstm (LSTM) (None, 18, 45) 8460 _________________________________________________________________ dropout (Dropout) (None, 18, 45) 0 _________________________________________________________________ lstm_1 (LSTM) (None, 18, 45) 16380 _________________________________________________________________ dropout_1 (Dropout) (None, 18, 45) 0 _________________________________________________________________ time_distributed (TimeDistri (None, 18, 1) 46 ================================================================= Total params: 24,886 Trainable params: 24,886 Non-trainable params: 0 _________________________________________________________________
history = model.fit(X_train, y_train, epochs=EPOCHS, batch_size=BATCHSIZE, validation_split=VALIDATIONSPLIT,
callbacks=[keras.callbacks.EarlyStopping(monitor='val_loss', patience=5, mode='min')], shuffle=False)
Epoch 1/43 83/83 [==============================] - 2s 20ms/step - loss: 0.6073 - mse: 0.7044 - rmse: 0.6400 - val_loss: 0.3867 - val_mse: 0.3473 - val_rmse: 0.4599 Epoch 2/43 83/83 [==============================] - 1s 12ms/step - loss: 0.5049 - mse: 0.5169 - rmse: 0.5519 - val_loss: 0.3428 - val_mse: 0.3078 - val_rmse: 0.4115 Epoch 3/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4740 - mse: 0.4757 - rmse: 0.5213 - val_loss: 0.3145 - val_mse: 0.2851 - val_rmse: 0.3794 Epoch 4/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4576 - mse: 0.4560 - rmse: 0.5036 - val_loss: 0.2950 - val_mse: 0.2713 - val_rmse: 0.3563 Epoch 5/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4473 - mse: 0.4438 - rmse: 0.4921 - val_loss: 0.2829 - val_mse: 0.2630 - val_rmse: 0.3405 Epoch 6/43 83/83 [==============================] - 1s 13ms/step - loss: 0.4399 - mse: 0.4351 - rmse: 0.4834 - val_loss: 0.2746 - val_mse: 0.2574 - val_rmse: 0.3283 Epoch 7/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4349 - mse: 0.4295 - rmse: 0.4772 - val_loss: 0.2685 - val_mse: 0.2533 - val_rmse: 0.3188 Epoch 8/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4304 - mse: 0.4249 - rmse: 0.4718 - val_loss: 0.2626 - val_mse: 0.2500 - val_rmse: 0.3102 Epoch 9/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4274 - mse: 0.4219 - rmse: 0.4681 - val_loss: 0.2586 - val_mse: 0.2475 - val_rmse: 0.3032 Epoch 10/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4241 - mse: 0.4191 - rmse: 0.4644 - val_loss: 0.2563 - val_mse: 0.2454 - val_rmse: 0.2978 Epoch 11/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4219 - mse: 0.4164 - rmse: 0.4616 - val_loss: 0.2528 - val_mse: 0.2436 - val_rmse: 0.2926 Epoch 12/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4203 - mse: 0.4141 - rmse: 0.4595 - val_loss: 0.2510 - val_mse: 0.2424 - val_rmse: 0.2888 Epoch 13/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4183 - mse: 0.4125 - rmse: 0.4575 - val_loss: 0.2486 - val_mse: 0.2412 - val_rmse: 0.2851 Epoch 14/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4173 - mse: 0.4111 - rmse: 0.4559 - val_loss: 0.2468 - val_mse: 0.2400 - val_rmse: 0.2819 Epoch 15/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4164 - mse: 0.4101 - rmse: 0.4548 - val_loss: 0.2457 - val_mse: 0.2393 - val_rmse: 0.2798 Epoch 16/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4152 - mse: 0.4089 - rmse: 0.4534 - val_loss: 0.2439 - val_mse: 0.2383 - val_rmse: 0.2771 Epoch 17/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4142 - mse: 0.4081 - rmse: 0.4527 - val_loss: 0.2426 - val_mse: 0.2377 - val_rmse: 0.2752 Epoch 18/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4140 - mse: 0.4076 - rmse: 0.4523 - val_loss: 0.2417 - val_mse: 0.2372 - val_rmse: 0.2737 Epoch 19/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4136 - mse: 0.4074 - rmse: 0.4522 - val_loss: 0.2408 - val_mse: 0.2369 - val_rmse: 0.2724 Epoch 20/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4125 - mse: 0.4064 - rmse: 0.4514 - val_loss: 0.2396 - val_mse: 0.2365 - val_rmse: 0.2710 Epoch 21/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4122 - mse: 0.4063 - rmse: 0.4512 - val_loss: 0.2406 - val_mse: 0.2366 - val_rmse: 0.2715 Epoch 22/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4122 - mse: 0.4061 - rmse: 0.4510 - val_loss: 0.2393 - val_mse: 0.2364 - val_rmse: 0.2701 Epoch 23/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4120 - mse: 0.4058 - rmse: 0.4508 - val_loss: 0.2394 - val_mse: 0.2362 - val_rmse: 0.2699 Epoch 24/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4117 - mse: 0.4054 - rmse: 0.4505 - val_loss: 0.2394 - val_mse: 0.2360 - val_rmse: 0.2697 Epoch 25/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4113 - mse: 0.4050 - rmse: 0.4499 - val_loss: 0.2378 - val_mse: 0.2355 - val_rmse: 0.2679 Epoch 26/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4112 - mse: 0.4045 - rmse: 0.4497 - val_loss: 0.2377 - val_mse: 0.2354 - val_rmse: 0.2677 Epoch 27/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4110 - mse: 0.4050 - rmse: 0.4496 - val_loss: 0.2377 - val_mse: 0.2354 - val_rmse: 0.2675 Epoch 28/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4111 - mse: 0.4047 - rmse: 0.4495 - val_loss: 0.2363 - val_mse: 0.2351 - val_rmse: 0.2660 Epoch 29/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4107 - mse: 0.4047 - rmse: 0.4495 - val_loss: 0.2361 - val_mse: 0.2349 - val_rmse: 0.2658 Epoch 30/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4106 - mse: 0.4046 - rmse: 0.4491 - val_loss: 0.2356 - val_mse: 0.2348 - val_rmse: 0.2651 Epoch 31/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4104 - mse: 0.4040 - rmse: 0.4488 - val_loss: 0.2361 - val_mse: 0.2348 - val_rmse: 0.2654 Epoch 32/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4102 - mse: 0.4042 - rmse: 0.4485 - val_loss: 0.2351 - val_mse: 0.2345 - val_rmse: 0.2643 Epoch 33/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4099 - mse: 0.4038 - rmse: 0.4484 - val_loss: 0.2348 - val_mse: 0.2344 - val_rmse: 0.2639 Epoch 34/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4102 - mse: 0.4040 - rmse: 0.4486 - val_loss: 0.2344 - val_mse: 0.2344 - val_rmse: 0.2634 Epoch 35/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4096 - mse: 0.4037 - rmse: 0.4479 - val_loss: 0.2343 - val_mse: 0.2343 - val_rmse: 0.2631 Epoch 36/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4095 - mse: 0.4035 - rmse: 0.4480 - val_loss: 0.2338 - val_mse: 0.2339 - val_rmse: 0.2624 Epoch 37/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4097 - mse: 0.4037 - rmse: 0.4479 - val_loss: 0.2345 - val_mse: 0.2340 - val_rmse: 0.2630 Epoch 38/43 83/83 [==============================] - 1s 11ms/step - loss: 0.4098 - mse: 0.4039 - rmse: 0.4480 - val_loss: 0.2338 - val_mse: 0.2339 - val_rmse: 0.2622 Epoch 39/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4094 - mse: 0.4034 - rmse: 0.4475 - val_loss: 0.2331 - val_mse: 0.2337 - val_rmse: 0.2615 Epoch 40/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4091 - mse: 0.4034 - rmse: 0.4472 - val_loss: 0.2336 - val_mse: 0.2338 - val_rmse: 0.2619 Epoch 41/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4093 - mse: 0.4032 - rmse: 0.4473 - val_loss: 0.2329 - val_mse: 0.2336 - val_rmse: 0.2611 Epoch 42/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4091 - mse: 0.4028 - rmse: 0.4471 - val_loss: 0.2330 - val_mse: 0.2334 - val_rmse: 0.2609 Epoch 43/43 83/83 [==============================] - 1s 12ms/step - loss: 0.4089 - mse: 0.4029 - rmse: 0.4469 - val_loss: 0.2321 - val_mse: 0.2333 - val_rmse: 0.2599
import matplotlib.pyplot as plt
plt.plot(history.history['loss'], label='MAE Training loss')
plt.plot(history.history['val_loss'], label='MAE Validation loss')
plt.plot(history.history['mse'], label='MSE Training loss')
plt.plot(history.history['val_mse'], label='MSE Validation loss')
plt.plot(history.history['rmse'], label='RMSE Training loss')
plt.plot(history.history['val_rmse'], label='RMSE Validation loss')
plt.legend();
X_train_pred = model.predict(X_train, verbose=0)
train_mae_loss = np.mean(np.abs(X_train_pred - X_train), axis=1)
plt.hist(train_mae_loss, bins=50)
plt.xlabel('Train MAE loss')
plt.ylabel('Number of Samples');
def evaluate_prediction(predictions, actual, model_name):
errors = predictions - actual
mse = np.square(errors).mean()
rmse = np.sqrt(mse)
mae = np.abs(errors).mean()
print(model_name + ':')
print('Mean Absolute Error: {:.4f}'.format(mae))
print('Root Mean Square Error: {:.4f}'.format(rmse))
print('Mean Square Error: {:.4f}'.format(mse))
print('')
return mae,rmse,mse
mae,rmse,mse = evaluate_prediction(X_train_pred, X_train,"LSTM")
LSTM: Mean Absolute Error: 0.1885 Root Mean Square Error: 0.4379 Mean Square Error: 0.1917
model.save(MODELFILENAME+'.h5')
#càlcul del threshold de test
def calculate_threshold(X_test, X_test_pred):
distance = np.sqrt(np.mean(np.square(X_test_pred - X_test),axis=1))
"""Sorting the scores/diffs and using a 0.80 as cutoff value to pick the threshold"""
distance.sort();
cut_off = int(0.95 * len(distance));
threshold = distance[cut_off];
return threshold
for col in columns:
print ("####################### "+col +" ###########################")
#Standardize the test data
scaler = StandardScaler()
test_cpy = test.copy()
test[col] = scaler.fit_transform(test[[col]])
#creem seqüencia amb finestra temporal per les dades de test
X_test1, y_test1 = create_sequences(test[[col]], test[col])
print(f'Testing shape: {X_test1.shape}')
#evaluem el model
eval = model.evaluate(X_test1, y_test1)
print("evaluate: ",eval)
#predim el model
X_test1_pred = model.predict(X_test1, verbose=0)
evaluate_prediction(X_test1_pred, X_test1,"LSTM")
#càlcul del mae_loss
test1_mae_loss = np.mean(np.abs(X_test1_pred - X_test1), axis=1)
test1_rmse_loss = np.sqrt(np.mean(np.square(X_test1_pred - X_test1),axis=1))
# reshaping test prediction
X_test1_predReshape = X_test1_pred.reshape((X_test1_pred.shape[0] * X_test1_pred.shape[1]), X_test1_pred.shape[2])
# reshaping test data
X_test1Reshape = X_test1.reshape((X_test1.shape[0] * X_test1.shape[1]), X_test1.shape[2])
threshold_test = calculate_threshold(X_test1Reshape,X_test1_predReshape)
test1_score_df = pd.DataFrame(test[TIME_STEPS:])
test1_score_df['loss'] = test1_rmse_loss.reshape((-1))
test1_score_df['threshold'] = threshold_test
test1_score_df['anomaly'] = test1_score_df['loss'] > test1_score_df['threshold']
test1_score_df[col] = test[TIME_STEPS:][col]
#gràfic test lost i threshold
fig = go.Figure()
fig.add_trace(go.Scatter(x=test1_score_df.index, y=test1_score_df['loss'], name='Test loss'))
fig.add_trace(go.Scatter(x=test1_score_df.index, y=test1_score_df['threshold'], name='Threshold'))
fig.update_layout(showlegend=True, title='Test loss vs. Threshold')
fig.show()
#Posem les anomalies en un array
anomalies1 = test1_score_df.loc[test1_score_df['anomaly'] == True]
anomalies1.shape
print('anomalies: ',anomalies1.shape); print();
#Gràfic dels punts i de les anomalíes amb els valors de dades transformades per verificar que la normalització que s'ha fet no distorssiona les dades
fig = go.Figure()
fig.add_trace(go.Scatter(x=test1_score_df.index, y=scaler.inverse_transform(test1_score_df[col]), name=col))
fig.add_trace(go.Scatter(x=anomalies1.index, y=scaler.inverse_transform(anomalies1[col]), mode='markers', name='Anomaly'))
fig.update_layout(showlegend=True, title='Detected anomalies')
fig.show()
print ("######################################################")
####################### PM1 ########################### Testing shape: (762, 18, 1) 24/24 [==============================] - 0s 4ms/step - loss: 0.4733 - mse: 0.7910 - rmse: 0.5344 evaluate: [0.4733470678329468, 0.7909653186798096, 0.5344359874725342]
<ipython-input-17-48420fb1aa44>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy test[col] = scaler.fit_transform(test[[col]])
LSTM: Mean Absolute Error: 0.1845 Root Mean Square Error: 0.5936 Mean Square Error: 0.3524
anomalies: (60, 10)
###################################################### ####################### PM25 ########################### Testing shape: (762, 18, 1) 1/24 [>.............................] - ETA: 0s - loss: 0.9245 - mse: 2.5176 - rmse: 1.02
<ipython-input-17-48420fb1aa44>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
24/24 [==============================] - 0s 3ms/step - loss: 0.4898 - mse: 0.7379 - rmse: 0.5506 evaluate: [0.48981156945228577, 0.7378596663475037, 0.5505625605583191] LSTM: Mean Absolute Error: 0.1964 Root Mean Square Error: 0.5526 Mean Square Error: 0.3053
anomalies: (93, 10)
###################################################### ####################### PM10 ########################### Testing shape: (762, 18, 1) 1/24 [>.............................] - ETA: 0s - loss: 0.7750 - mse: 1.1981 - rmse: 0.88
<ipython-input-17-48420fb1aa44>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
24/24 [==============================] - 0s 4ms/step - loss: 0.4992 - mse: 0.7139 - rmse: 0.5614 evaluate: [0.4992370307445526, 0.7138815522193909, 0.561396062374115] LSTM: Mean Absolute Error: 0.2026 Root Mean Square Error: 0.5158 Mean Square Error: 0.2660
anomalies: (43, 10)
###################################################### ####################### PM1ATM ########################### Testing shape: (762, 18, 1) 15/24 [=================>............] - ETA: 0s - loss: 0.5065 - mse: 0.7561 - rmse: 0.5816
<ipython-input-17-48420fb1aa44>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
24/24 [==============================] - 0s 3ms/step - loss: 0.5076 - mse: 0.7461 - rmse: 0.5734 evaluate: [0.5076465606689453, 0.7460874319076538, 0.5734124183654785] LSTM: Mean Absolute Error: 0.1981 Root Mean Square Error: 0.5108 Mean Square Error: 0.2609
anomalies: (61, 10)
###################################################### ####################### PM25ATM ########################### Testing shape: (762, 18, 1) 15/24 [=================>............] - ETA: 0s - loss: 0.5021 - mse: 0.7688 - rmse: 0.5765
<ipython-input-17-48420fb1aa44>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
24/24 [==============================] - 0s 4ms/step - loss: 0.5025 - mse: 0.7511 - rmse: 0.5675 evaluate: [0.5024616718292236, 0.751102864742279, 0.567493736743927] LSTM: Mean Absolute Error: 0.1961 Root Mean Square Error: 0.5224 Mean Square Error: 0.2729
anomalies: (60, 10)
###################################################### ####################### PM10ATM ########################### Testing shape: (762, 18, 1) 16/24 [===================>..........] - ETA: 0s - loss: 0.4673 - mse: 0.6684 - rmse: 0.53
<ipython-input-17-48420fb1aa44>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
24/24 [==============================] - 0s 3ms/step - loss: 0.4961 - mse: 0.7195 - rmse: 0.5578 evaluate: [0.49610012769699097, 0.7195131778717041, 0.5578391551971436] LSTM: Mean Absolute Error: 0.2008 Root Mean Square Error: 0.5271 Mean Square Error: 0.2779
anomalies: (59, 10)
######################################################